The objective is to design, implement, and clinically assess ETSense, an adaptive, compact, portable essential tremor (ET) monitor for optimizing therapeutic interventions. ET is characterized primarily by postural and kinetic (action) tremors of the limbs, which are rated by various subjective tremor rating scales. These scales all provide a discrete, subjective symptom rating at a discrete point in time, require a clinician to visually assess the patient, and cannot capture complex fluctuations that occur throughout the day in response to interventions. Objectively capturing ET symptoms continuously during daily activities and using adaptive algorithms to both classify tremor types and severity will help clinicians better titrate therapy to minimize symptom fluctuations and expand care to rural and underserved populations. The Phase I ETSense effort successfully used kinematic data recorded from a sensor unit placed on the finger of subjects with ET to discriminate tremor from voluntary motion associated with daily activities and objectively quantified tremor severity with scores highly correlated with clinicians' qualitative ratings, providing a standardized platform for continuous ET assessment. Tremor quantification algorithms were extrapolated to non-standardized tasks, suggesting that it is feasible to rate tremor continuously throughout the day during activities of daily living. The three primary innovations of the proposed system include: 1) a compact, portable, user-worn device for continuous monitoring during ADLs, 2) intelligent, adaptive algorithms to continuously classify tremor type and rate severity, and 3) web-based access to symptom response reports. The clinically deployable system will be contained in a lightweight, finger-worn housing for continuous wear while patients perform everyday tasks at home or in public. A push button diary will allow the patient to indicate when medication is taken. All data will be stored in memory for subsequent analysis and report generation detailing symptom fluctuations in response to therapeutic interventions. Adaptive algorithms developed in Phase I will be further optimized to account for voluntary motion that can create tremor false positives or mask over kinematic tremor signals. The system will shift between scoring algorithms (i.e. rest, kinetic) based on any voluntary motion detected. After data collection is complete, clinicians will use a web-interface to view patient reports. These reports will detail tremor type, severity, and fluctuations, as well as when medication was taken to aid clinicians in optimizing existing therapeutic interventions or in the research and development of novel treatment protocols. We hypothesize that the commercial ETSense system will 1) continuously quantify tremor severity throughout the day during activities of daily living, 2) improve patient outcomes with better and/or faster medication optimization, 3) decrease healthcare costs by reducing office visits, and 4) enable the testing and validation of novel therapeutic interventions, facilitated by high-resolution continuous home monitoring.